CN110429587A - A kind of two stages electrical network parameter estimation method - Google Patents
A kind of two stages electrical network parameter estimation method Download PDFInfo
- Publication number
- CN110429587A CN110429587A CN201910653860.8A CN201910653860A CN110429587A CN 110429587 A CN110429587 A CN 110429587A CN 201910653860 A CN201910653860 A CN 201910653860A CN 110429587 A CN110429587 A CN 110429587A
- Authority
- CN
- China
- Prior art keywords
- parameter
- suspicious
- wrong
- wrong parameter
- measuring section
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Measurement Of Resistance Or Impedance (AREA)
Abstract
The present invention relates to a kind of two stages method for parameter estimation, comprising the following steps: 1) establishes the suspicious wrong parameter detection model based on optimal measuring section, detect the suspicious wrong parameter in power grid；2) it using suspicious wrong parameter as augmented state state variable, establishes segmentation augmented state and estimates model, realize the amendment of suspicious wrong parameter.The present invention has many advantages, such as that anti-bad data interference, computational accuracy is high, has a extensive future compared with prior art.
Description
Technical field
The present invention relates to parameters of electric power system identification technique fields, more particularly, to a kind of two stages method for parameter estimation.
Background technique
With the continuous expansion of power grid, electric power system model is become more complicated, and the quantity of sorts of systems parameter is also exploded
Property increase, more stringent requirements are proposed for accuracy of the normal work of system to parameter.Simultaneously as periodical grid maintenance and
Transformation leads to the sporadic parameter drifts of artificial the setting mistake and various equipment of parameter, and cause based on system parameter advanced answers
With generating larger calculating error, or even do not restrain.Therefore, how accurately identifying and correcting to network parameter mistake is electric system
Major issue in Accurate Model.
Existing parameter identification method can substantially be divided into equation, measurement method, heuristic, single line modelling, sensitivity
Six class of analytic approach and generalized petri net model.
Equation.The basic thought of equation is to pass through line length, material according to the empirical equation in power system calculation handbook
The many factors COMPREHENSIVE CALCULATING such as matter.This calculating is usually calculated at the route initial stage of building up, since line parameter circuit value is by a variety of
The influence of topoclimate and ageing equipment, therefore the calculated value that empirically formula obtains generally is only applicable to power grid rule
It draws, is not able to satisfy requirement of the system real time execution to line parameter circuit value accuracy.
Measurement method.This method mainly carries out field survey to line parameter circuit value using measuring instrument, and measurement method is divided to two kinds:
It is another then be after measuring each phase parameter one is directly surveying positive sequence and Zero sequence parameter, then phase parameter is converted into sequence through phase sequence
Parameter.Such method usually requires to carry out power failure experiment or applies alien frequencies power supply, not can avoid complicated manual record and work
Amount, and the parameter drifted about can not be found in time.
Heuristic.Heuristic is to pass through observation state after particular step size change parameter on the basis of known suspicious parameter
Estimate the situation of change of index, so that it is determined that parameter is to owe amendment or cross to correct, is finally made by the adjustment of step-length
State estimation index preferably when the parameter value as finally estimated of corresponding parameter.Such method is although easy to operate, but it is most
Excellent amendment step-length is difficult to accurately obtain, and not can avoid redundancy of effort.
Single line modelling.This method is model simultaneously benefit to single line using the impedance of single line as unknown parameter
The identification of its parameter is realized with the Correlated Case with ARMA Measurement data of the route.The shortcomings that such method, is since Chinese power transmission network is not yet real
Existing PMU's is completely covered, and limits its engineer application.
Sensitivity Analysis Method.Basic thought is the spirit established between parameter error and error in measurement using state estimation result
Sensitive matrix, so that the size of parameter error can be obtained according to the estimation of error in measurement.Such side when wrong parameter is excessive
The Parameter Estimation Precision of method is highly susceptible to influencing each other between bad data and wrong parameter, will lead to parameter mistake when serious
Amendment.
Generalized petri net model.Its basic thought is to carry out state estimation for suspicious parameter as augmented state variable.Such side
Dimension of the method due to increasing quantity of state, so that the redundancy of system declines, numerical stability is reduced, and is even resulted in not when serious
Convergence.
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide a kind of " detections+identification "
Two stages parameter identification method.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of two stages method for parameter estimation, comprising the following steps:
1) the suspicious wrong parameter detection model based on optimal measuring section is established, detects the suspicious mistake ginseng in power grid
Number；
2) it using suspicious wrong parameter as augmented state state variable, establishes segmentation augmented state and estimates model, realization can
Doubt the amendment of wrong parameter.
The optimal measuring section detection model are as follows:
In formula, t is optimal measuring section number；PeFor the wrong parameter vector of N-dimensional network；N is all measuring section sums；
λ∑iFor the accumulation Lagrange multiplier of wrong parameter, its calculation formula is:
λ∑i=λI, 1+λI, 2+L+λI, N
In formula, xiFor the quantity of state of i-th of measuring section；Ji(xi, Pe) be i-th of measuring section objective function；λI, 1+
λI, 2+L+λI, NFor vector λiIn each element；It is measurement equation to error in measurement PeRefined lattice than matrix transposition.
The segmentation augmented state estimates model are as follows:
In, hi(x, Pe) it is the measurement accounting equation containing state variable and augmented state variable；K is iterative step；ziIt is
I measurement, ωiFor its weight；X is the state variable vector that voltage magnitude and phase angle are constituted；σiIt is poor for measuring standard；For amount
Survey window width.
Compared with prior art, the invention has the following advantages that
One, computational accuracy is high: the present invention by segmented objects function model, reduce influencing each other between wrong parameter and
Reduce influence of the bad data to Parameter Estimation Precision, there is very high Parameter Estimation Precision；
Two, it the ability with very strong anti-bad data interference: is deposited compared to other parameter identification methods in bad data
When, the accuracy decline of parameter identification does not restrain even, and the present invention can be existed simultaneously in bad data and multiple parameters mistake
The detection and amendment of Shi Shixian parameter error；
Three, have a extensive future: suspicious wrong parameter detection method through the invention can will face in conjunction with artificial experience
The grid equipment parameter of nearly life cycle is also used as suspicious wrong parameter, to avoid the missing inspection of wrong parameter well；It is logical
The segment processing of state estimation model is crossed, can automatically be eliminated between wrong parameter, between wrong parameter and mistake measurement
It influences each other, it is simple and convenient.Therefore the present invention has a good application prospect.
Detailed description of the invention
Fig. 1 is flow chart of the method for the present invention.
Specific embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.
As shown in Figure 1, the present invention provides a kind of two stages method for parameter estimation, including the following steps executed in order:
Step 1) establishes the suspicious wrong parameter detection model based on optimal measuring section, detects the suspicious mistake in power grid
Miss parameter:
In formula, t is optimal measuring section number；PeFor the wrong parameter vector of N-dimensional network；N is all measuring section sums；
λ∑iFor the accumulation Lagrange multiplier of wrong parameter, its calculation formula is:
λ∑i=λI, 1+λI, 2+L+λI, N
In formula, xiFor the quantity of state of i-th of measuring section；Ji(xiPe) be i-th of measuring section objective function；λI, 1+
λI, 2+L+λI, NFor vector λiIn each element；It is measurement equation to error in measurement PeRefined lattice than matrix transposition.
Step 2) establishes segmentation augmented state and estimates model using suspicious wrong parameter as augmented state state variable, real
The amendment of existing suspicious wrong parameter:
In, hi(x, Pe) it is the measurement accounting equation containing state variable and augmented state variable；K is iterative step；ziIt is
I measurement, ωiFor its weight；X is the state variable vector that voltage magnitude and phase angle are constituted；σiIt is poor for measuring standard；For amount
Survey window width.
Two stages method for parameter estimation embodiment
This paper example has worked out corresponding program under Microsoft Visual C++2010 environment, using IEEE9 node
System, verifies the feasibility and validity of proposed method for correcting and recognize branch impedance, and all amounts of example are adopted
It is indicated with per unit value.
9 node examples are measured using configuration completely, and metric data uses calculation of tidal current.The bad data of setting is in tide
3 times of standard deviations of the measurement are added on the basis of stream calculation result.Table 1- table 4 the result shows that, methods herein to reactance estimate
More sensitive, the convergence precision of resistance estimation and reactance estimation is set as 0.001 and 0.0001.
Transformer active is provided in table 1 (a)With it is idleAnd the active P of route45Three measurement mistakes.By table 1
(b) it is found that when, there are when bad data, especially working as bad data in systemWhen with wrong parameter (route 2-7) strong correlation, ginseng
The influence that the result of number identification will receive bad data leads to accuracy decline, or even does not restrain (by table 4 (a) and table 4 (b)), but
It is that the convergence rate of this method can complete convergence not being affected and need not make to bad data extra process.As a result
Show influence of the non-strong correlation measurement of the elimination that the present invention can be adaptive to wrong parameter, directly obtains the knot of parameter identification
Fruit measures without first distinguishing and rejecting mistake caused by bad data and parameter error.But when bad data and parameter
When mistake strong correlation, it may cause accuracy decline or even do not restrain (by table 1, table 4), this point needs further research.
The setting of table 1 (a) strong correlation bad data
Table 1 (b) strong correlation bad data parameter Estimation
| Route | True value | Setting value | Detection | Estimated value |
| L2-7 | 0+0.0625j | 0+0.0645j | L2-7 | -0.000003+0.062660j |
The non-strong correlation bad data setting of table 2 (a)
The non-strong correlation bad data parameter Estimation of table 2 (b)
| Route | True value | Setting value | Detection | Estimated value |
| L2-7 | 0+0.0625j | 0+0.0645j | L2-7 | -0.000002+0.062503 |
The non-strong correlation bad data setting of table 3 (a) (r cannot ignore)
The non-strong correlation bad data resistance estimation of table 3 (b)
(r cannot ignore) is arranged in table 4 (a) strong correlation bad data
Table 4 (b) strong correlation bad data resistance identification result
| Route | True value | Setting value | Detection | Estimated value |
| r45 | 0.01 | 0.011 | r45 | It does not restrain |
Claims (3)
1. a kind of two stages power grid parameter identification method, which comprises the following steps:
1) the suspicious wrong parameter detection model based on optimal measuring section is established, detects the suspicious wrong parameter in power grid；
2) it using suspicious wrong parameter as augmented state state variable, establishes segmentation augmented state and estimates model, realize suspicious mistake
The accidentally amendment of parameter.
2. suspicious wrong parameter according to claim 1 detects detection method, which is characterized in that the optimal measurement is disconnected
Face detection model are as follows:
In formula, t is optimal measuring section number；PeFor the wrong parameter vector of N-dimensional network；N is all measuring section sums；λ∑iFor
The accumulation Lagrange multiplier of wrong parameter, its calculation formula is:
λ∑i=λI, 1+λI, 2+L+λI, N
In formula, xiFor the quantity of state of i-th of measuring section；Ji(xi, Pe) be i-th of measuring section objective function；λI, 1+λI, 2+L
+λI, NFor vector λiIn each element；It is measurement equation to error in measurement PeRefined lattice than matrix transposition.
3. suspicious wrong parameter estimation method according to claim 2, which is characterized in that the segmentation augmented state is estimated
Count model are as follows:
In, hi(x, Pe) it is the measurement accounting equation containing state variable and augmented state variable；K is iterative step；ziIt is i-th
It measures, ωiFor its weight；X is the state variable vector that voltage magnitude and phase angle are constituted；σiIt is poor for measuring standard；To measure
Window width.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910653860.8A CN110429587A (en) | 2019-07-19 | 2019-07-19 | A kind of two stages electrical network parameter estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910653860.8A CN110429587A (en) | 2019-07-19 | 2019-07-19 | A kind of two stages electrical network parameter estimation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110429587A true CN110429587A (en) | 2019-11-08 |
Family
ID=68410122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910653860.8A Withdrawn CN110429587A (en) | 2019-07-19 | 2019-07-19 | A kind of two stages electrical network parameter estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110429587A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008253076A (en) * | 2007-03-30 | 2008-10-16 | Hitachi Ltd | Method for estimating state of power system |
| CN102280877A (en) * | 2011-07-25 | 2011-12-14 | 清华大学 | Method for identifying parameter of poor branch of power system through a plurality of measured sections |
| CN104836223A (en) * | 2014-11-14 | 2015-08-12 | 浙江大学 | Power grid parameter error and bad data coordinated identification and estimation method |
| CN105406471A (en) * | 2015-12-23 | 2016-03-16 | 云南电力调度控制中心 | Bad data identification and estimation method for power grid |
| CN110048402A (en) * | 2018-12-31 | 2019-07-23 | 国网辽宁省电力有限公司 | A kind of two stages electrical network parameter estimation method |
-
2019
- 2019-07-19 CN CN201910653860.8A patent/CN110429587A/en not_active Withdrawn
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008253076A (en) * | 2007-03-30 | 2008-10-16 | Hitachi Ltd | Method for estimating state of power system |
| CN102280877A (en) * | 2011-07-25 | 2011-12-14 | 清华大学 | Method for identifying parameter of poor branch of power system through a plurality of measured sections |
| CN104836223A (en) * | 2014-11-14 | 2015-08-12 | 浙江大学 | Power grid parameter error and bad data coordinated identification and estimation method |
| CN105406471A (en) * | 2015-12-23 | 2016-03-16 | 云南电力调度控制中心 | Bad data identification and estimation method for power grid |
| CN110048402A (en) * | 2018-12-31 | 2019-07-23 | 国网辽宁省电力有限公司 | A kind of two stages electrical network parameter estimation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10802056B2 (en) | Updating a topology of a distribution network by successive reallocation of the meters | |
| CN101635457B (en) | Electric network parameter estimation method based on parameter sensitivity of state estimation residual error | |
| CN107016236B (en) | Power grid false data injection attack detection method based on nonlinear measurement equation | |
| CN103944165B (en) | A kind of bulk power grid parameter identification method of estimation | |
| CN104836223B (en) | Power grid parameter error and bad data coordinated identification and estimation method | |
| CN106646140B (en) | HVDC transmission line internal fault external fault recognition methods based on measurement wave impedance | |
| CN102175922B (en) | Phasor measurement unit (PMU) measurement data-based power line parameter identification and estimation method | |
| CN111144435A (en) | Electric energy abnormal data monitoring method based on LOF and verification filtering framework | |
| CN107453484B (en) | SCADA data calibration method based on WAMS information | |
| CN111950396B (en) | Meter reading neural network identification method | |
| CN107658881A (en) | Voltage stability critical point determination methods based on Thevenin's equivalence method | |
| CN110190600A (en) | It is a kind of to measure the three-phase distribution net topology discrimination method that neighbour returns based on AMI | |
| CN110518590A (en) | Consider the linear tidal current computing method of power distribution network of static load characteristics | |
| CN113484819B (en) | Method for diagnosing metering faults of electric energy meter in limited range based on high-frequency current sampling | |
| CN108448585A (en) | A kind of electric network swim equation solution method of linearization based on data-driven | |
| CN109753762A (en) | Based on the modified power distribution network two stages network topology identification method of classification and device | |
| CN110133444A (en) | A kind of Fault Locating Method based on positive sequence voltage variable quantity, apparatus and system | |
| CN103324858A (en) | Three-phase load flow state estimation method of power distribution network | |
| CN102636706B (en) | Method for identifying branches with parameter errors in power grid | |
| CN106372440B (en) | A kind of adaptive robust state estimation method of the power distribution network of parallel computation and device | |
| CN110048402A (en) | A kind of two stages electrical network parameter estimation method | |
| CN113189513A (en) | Ripple-based redundant power supply current sharing state identification method | |
| CN110429587A (en) | A kind of two stages electrical network parameter estimation method | |
| CN109638811B (en) | Power distribution network voltage power sensitivity robust estimation method based on model equivalence | |
| CN109193639B (en) | Robust estimation method for power system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20191108 |